For over a century, resistance of radiation to living tissues has been associated with hypoxia, a local lack of molecular oxygen, or low pO?. The focus of the past funding cycle has been to validate the hypothesis that has been assumed, but not proven over the past century, that treatment focused specifically on regions of tumors with low pO?, voxels less than 10 torr, hypoxia boosts, would improve tumor curability. This research grant has used Electron Paramagnetic Resonance (EPR) imaging to provide absolute pO? images in volume elements or voxels of murine tumors with 1 torr pO? resolution and 0.5 mm spatial resolution in FSa carcinomas in the legs of C?H mice. The voxel pO? correlates with local Oxylite measurements. EPR pO? image based hypoxic fractions, HF10 (fraction of tumor voxels with pO? less than 10 torr) correlates with hypoxia proteins VEGF, CAIX, and HIF1?, and with the curability of tumors given a dose of radiation sufficient to cure 50% of 450 l tumors (TCD50). This established EPR pO? imaging as a reliable locator of relevant radiobiologically relevant hypoxia. To determine if pO? based dose painting improves tumor cure, we implemented the XRAD???Cx system to deliver gantry based x-ray treatments to mouse tumors accurately registered with EPR pO2 images. We implemented rapid 3D printing Tungsten loaded, conformal plastic blocks to compare treating ~100% of hypoxic tumor voxels with hypoxia avoidance. Only then did we observe significant (p=0.02) tumor control differences between hypoxic boosts and anti-boosts. This is the first validation of hypoxia based dose painting in mammalian tumors. The systematics of the differences between hypoxic and normal pO? tumor tissue now need to be determined in several animal models with different immunologic conditions and with fractionation to guide eventual human use, possibly based on reductive retention of ?F-nitroimidazole PET images. We propose the following program investigating the systematics of EPR pO? image based dose painting: 1) Determine the in vivo hypoxic and separate normally oxygenated tumor tissue pO? control doses (TCD??Hypox and TCD??Ox), an in vivo oxygen enhancement ratios (OER) for orthotopic FSa and RIF1 fibrosarcomas, MCa4 mammary carcinomas grown orthotopically and, to determine the immunogenic status dependence, in human PC3 prostate carcinoma xenografts in athymic nude mice. 2) Determine the influence of three dose fractionation on hypoxic tumor control and oxygen enhancement ratios (?fTCD??Hypox, ?fTCD??Ox: ?fOER) These experiments will provide ranges of in vivo variation from which to estimate in vivo oxygen enhancement ratios to guide early human trials of hypoxic boost/dose painting treatment. The success of preclinical determination of OER in multiple model tumors may suggest means by which to correct PET based human hypoxic tumor imaging. We also show technology suggesting EPR imaging in human subjects.